本数据集提供了1970-2020年逐日关键积雪属性数据集。积雪属性包括积雪范围、积雪深度、雪水当量，空间分辨率为5km，时间范围为1970-2020年积雪期（当年10月1日至次年4月30日）。其中，1980-2020年积雪范围数据使用AVHRR表面反射率数据、Landsat-5TM数据，结合地面雪深观测数据、中国长期日积雪深度数据集、地表温度和DEM等数据源，采用改进的云检测算法、多级积雪判别算法与间隙填补策略构建；1980-2020年积雪深度数据融合了最新校准的增强分辨率亮度温度与光学积雪面积比例和积雪覆盖日数等数据，基于深度学习FT-Transformer模型反演获取；雪水当量数据是通过逐月平均的积雪密度数据将雪深转化为5 km空间分辨率的逐日雪水当量数据。采用LightGBM（Light Gradient Boosting Machine）模型，以 1980-2020 年雪深数据为目标变量，结合气温、降水、风速、DEM、坡度、坡向正弦值、坡向余弦值和植被类型等，生成1970-1979年逐日5km空间分辨率雪深重建数据；利用1970-1979年逐月积雪密度数据，将雪深数据转化为5 km空间分辨率的1970-1979年逐日雪水当量数据产品；同时，利用1980-2020年积雪空间分布数据，确定有雪像元在积雪深度产品上的雪深平均阈值，识别有雪和无雪像元，完成了1970-1979年逐日5 km积雪范围产品。数据精度上，1980-2020积雪范围：使用混淆矩阵以及四种准确率指标包括总体精度（OA）、生产者精度（PA）、用户精度（UA）和Kappa 系数来评估本产品，OA范围在80%–90%之间，PA和UA范围在70%–90%，CK值范围为0.61至0.8。使用38年的CMA地面积雪深度测量191个站点验证本产品，大多气象站OA较高，普遍在80%–90%之间，但 PA、UA和CK的值较低。使用9张Landsat-5积雪覆盖面积图来进一步评估本产品，OA高达87.3%，UA较高和PA较低表明该产品在一定程度上存在低估积雪覆盖范围的倾向，Kappa值为0.695，这一数值也接近地面观测验证的kappa值（0.717）。1980-2020积雪深度：采用均方根误差（RMSE）、平均绝对误差（MAE）和相关系数（R）三个标准指标表示反演的雪深误差，利用2000-2020年的地面实测雪深进行评估。验证结果表明，三江源地区的5 km雪深的RMSE位于8~8.5 cm、MAE位于5.6~6.5 cm，R大于0.7,相较于中国长时间序列的雪深数据（25 km）（RMSE位于10~11.5 cm、MAE位于7.5~8.3 cm，R大于0.45）具有更优的精度。1970-1979积雪深度和雪水当量：采用测试集均方根误差（RMSE）、平均绝对误差（MAE）和相关系数（R）三个指标评价模型精度。结果显示，积雪深度测试集均方根误差为3.94cm，MAE为2.51cm，R为0.51。雪水当量的评估精度为RMSE为5.12mm，MAE为3.26mm。", "ds_source": "", "ds_process_way": "", "ds_quality": "", "ds_acq_start_time": "1970-10-01 00:00:00", "ds_acq_end_time": "2020-04-30 00:00:00", "ds_acq_place": "三江源地区", "ds_acq_lon_east": 103.0, "ds_acq_lat_south": 31.0, "ds_acq_lon_west": 88.0, "ds_acq_lat_north": 38.0, "ds_acq_alt_low": null, "ds_acq_alt_high": null, "ds_share_type": "open-access", "ds_total_size": 1472007515, "ds_files_count": 0, "ds_format": "*.tif", "ds_space_res": "5km", "ds_time_res": "日", "ds_coordinate": "无", "ds_projection": "", "ds_thumbnail": "2001238a-ea66-4e31-90bd-e8e71cf45633.png", "ds_thumb_from": 0, "ds_ref_way": "", "paper_ref_way": "", "ds_ref_instruction": "None", "ds_from_station": null, "organization_id": "52b7b79b-860c-49a5-9083-9a70cf8bed5a", "ds_serv_man": null, "ds_serv_phone": null, "ds_serv_mail": null, "doi_value": "", "subject_codes": [ "170" ], "quality_level": 0, "publish_time": "2026-09-01 15:37:42", "last_updated": "2026-09-01 15:37:42", "protected": false, "protected_to": null, "lang": "zh", "cstr": "11738.11.ncdc.db7701.2026", "i18n": { "en": { "title": "Long-term key daily snow parameter dataset for the Three-River Source region from 1970 to 2020", "ds_format": "*.tif", "ds_source": "", "ds_quality": "", "ds_ref_way": "", "ds_abstract": "
This dataset provides daily data on key snow cover parameters from 1970 to 2020. Snow cover parameters include snow cover extent (SCE), snow depth (SD), and snow water equivalent (SWE). The spatial resolution is 5km, and the snow cover period is from 1970 to 2020 (October 1 of the current year to April 30 of the following year). SCE data from 1980 to 2020 uses AVHRR surface reflectance data and Landsat-5TM data, combined with ground snow depth observation data, long-term daily snow depth observations, surface temperature and DEM, and is constructed using improved cloud detection algorithms, multi-level snow cover discrimination algorithms and gap filling strategies; The 1980-2020 snow depth data combines the newly calibrated enhanced resolution brightness temperature to optical snow area ratio and snow cover days. It is retrieved using the deep learning FT-Transformer model and calculated from monthly average snow density. SD is converted into SWE data with a spatial resolution of 5 km. Using LightGBM The (Light Gradient Boosting Machine) model takes snow depth data from 1980 to 2020 as the target variable, combines air temperature, precipitation, wind speed, DEM, slope, sine values of slope direction, cosine values of slope direction and vegetation type, etc., to generate daily SD reconstruction data with 5km spatial resolution from 1970 to 1979; Using the monthly snow density data from 1970 to 1979, SD was transformed into a daily SWE product from 1970 to 1979 with a spatial resolution of 5 km; at the same time, using the snow cover spatial distribution data from 1980 to 2020, the average SD threshold of snow-containing pixels on SD product was determined, snow and snow-free pixels were identified, and the daily 5-kilometer snow cover range product from 1970 to 1979 was completed.", "ds_time_res": "day", "ds_acq_place": "Three‑River‑Source Region", "ds_space_res": "5 km", "ds_projection": "", "ds_share_type": "open-access", "ds_process_way": "", "ds_ref_instruction": "" } }, "submit_center_id": "ncdc", "data_level": 0, "recommendation_value": 0, "license_type": "https://creativecommons.org/licenses/by/4.0/", "doi_reg_from": "reg_local", "cstr_reg_from": "reg_local", "doi_not_reg_reason": null, "cstr_not_reg_reason": null, "is_paper_in_submitting": false, "belong_to_nieer": false, "allow_update_data": false, "created_from": "fair", "ds_topic_tags": [ "积雪范围", "雪深", "雪水当量", "长时序", "机器学习" ], "ds_subject_tags": [ "地球科学" ], "ds_class_tags": [], "ds_locus_tags": [ "三江源" ], "ds_time_tags": [ 1970, 1971, 1972, 1973, 1974, 1975, 1976, 1977, 1978, 1979, 1980, 1981, 1982, 1983, 1984, 1985, 1986, 1987, 1988, 1989, 1990, 1991, 1992, 1993, 1994, 1995, 1996, 1997, 1998, 1999, 2000, 2001, 2002, 2003, 2004, 2005, 2006, 2007, 2008, 2009, 2010, 2011, 2012, 2013, 2014, 2015, 2016, 2017, 2018, 2019, 2020 ], "ds_contributors": [ { "true_name": "赵子胜", "email": "zhaozisheng@nieer.ac.cn", "work_for": "中国科学院西北生态环境资源研究院", "country": "中国" }, { "true_name": "杨瑞敏", "email": "yangruimin@lzufe.edu.cn", "work_for": "兰州财经大学", "country": "中国" }, { "true_name": "郝晓华", "email": "haoxh@lzb.ac.cn", "work_for": "中国科学院西北生态环境资源研究院", "country": "中国" }, { "true_name": "王海东", "email": "wanghd@lzufe.edu.cn", "work_for": "兰州财经大学", "country": "中国" }, { "true_name": "李杭璇", "email": "2023521174@link.tyut.edu.cn", "work_for": "太原理工大学", "country": "中国" }, { "true_name": "钟歆玥", "email": "xyzhong@lzb.ac.cn", "work_for": "中国科学院西北生态环境资源研究院", "country": "中国" }, { "true_name": "吴晓东", "email": "wuxd@lzb.ac.cn", "work_for": "中国科学院西北高原生物研究所", "country": "中国" } ], "ds_meta_authors": [ { "true_name": "赵子胜", "email": "zhaozisheng@nieer.ac.cn", "work_for": "中国科学院西北生态环境资源研究院", "country": "中国" }, { "true_name": "杨瑞敏", "email": "yangruimin@lzufe.edu.cn", "work_for": "兰州财经大学", "country": "中国" } ], "ds_managers": [ { "true_name": "郝晓华", "email": "haoxh@lzb.ac.cn", "work_for": "中国科学院西北生态环境资源研究院", "country": "中国" }, { "true_name": "钟歆玥", "email": "xyzhong@lzb.ac.cn", "work_for": "中国科学院西北生态环境资源研究院", "country": "中国" } ], "category": "积雪" }